Beyond Words: ESC‐Net Revolutionizes VQA by Elevating Visual Features and Defying Language Priors
Computational Intelligence · 2024-12 · 40 (6)
DOI: 10.1111/coin.70010 ↗Kurz gefasst
Fachartikel in der Zeitschrift Computational Intelligence, online erschienen am 03.12.2024. Verfasst von Souvik Chowdhury und 1 weiteren Autor:innen. Die Arbeit wurde bisher 41-mal zitiert (Crossref) und nennt 45 Literaturangaben.
Abstract (Originalsprache)
Language prior is a pressing problem in the VQA domain where a model provides an answer favoring the most frequent related answer. There are some methods that are adopted to mitigate language prior issue, for example, ensemble approach, the balanced data approach, the modified evaluation strategy, and the modified training framework. In this article, we propose a VQA model, “Ensemble of Spatial and Channel Attention Network (ESC‐Net),” to overcome the language bias problem by improving the visual features. In this work, we have used regional and global image features along with an ensemble of combined channel and spatial attention mechanisms to improve visual features. The model is a simpler and effective solution than existing methods to solve language bias. Extensive experiment show a remarkable performance improvement of 18% on the VQACP v2 dataset with a comparison to current state‐of‐the‐art (SOTA) models.
Zitieren
APA
Chowdhury, S., & Soni, B. (2024). Beyond Words: ESC‐Net Revolutionizes VQA by Elevating Visual Features and Defying Language Priors. Computational Intelligence, 40(6). https://doi.org/10.1111/coin.70010
Vancouver
Chowdhury S, Soni B. Beyond Words: ESC‐Net Revolutionizes VQA by Elevating Visual Features and Defying Language Priors. Computational Intelligence. 2024;40(6). doi:10.1111/coin.70010
BibTeX
@article{Chowdhury2024,
author = {Chowdhury, Souvik and Soni, Badal},
title = {Beyond Words: ESC‐Net Revolutionizes VQA by Elevating Visual Features and Defying Language Priors},
journal = {Computational Intelligence},
year = {2024},
volume = {40},
number = {6},
doi = {10.1111/coin.70010}
}
EUVIMED